Urology Journal UNRC/IUA Vol. 1, No. 4, 246-249 Autumn 2004 Printed in IRAN 246 O R I G I N A L A R T I C L E S Urological Oncology Correlation between Prostate Needle Biopsy and Radical Prostatectomy Gleason Gradings of 111 Cases with Prostatic Adenocarcinoma TAVANGAR SM1*, RAZI A2, MASHAYEKHI R1 1Department of Pathology, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran 2Department of Urology, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran ABSTRACT Purpose: There are conflicting reports in the literature about correlation of biopsy and prostatectomy Gleason scores in prostate carcinoma. The goal of this study was to determine the correlation of grading in these two types of pathologic materials. Materials and methods: The coupled Hematoxylin and Eosin slides of 111 patients with prostate carcinoma were collected. Gleason scores were determined. Patients who had undergone any therapy except surgery were excluded from the study. Correlation between grades was calculated by determination of correlation coefficient. Accuracy of biopsy grading in prediction of final grade was also determined by measuring the sen- sitivity, specificity, and positive and negative predictive values. Results: In 50 cases (45%), grade was underestimated in the biopsy. After dividing the cases into Gleason scores of 2 to 4, 5 to 6, 7, and 8 to 10, the most of under- graded cases (84.2%) were in the first group (Gleason score 2 to 4) and this rate reached 5% in the fourth group (Gleason score 8 to 10). The correlation coefficient measured was 0.535 in grade to grade comparing and 0.514 in group to group com- parison of the specimens. In low-grade tumors, grading in biopsy, in spite of high sen- sitivity (90.9%), had low positive predictive value (26.3%). Conclusion: There is a moderate direct linear relationship between scores in biopsy and prostatectomy specimens. But there is a high probability of underestimation of real Gleason score of the radical prostatectomy specimen in low-grade tumors. Pathologists and urologists must consider the phenomenon of undergrading in report- ing prostate specimens and managing patients. KEY WORDS: Gleason grading, needle biopsy, adenocarcinoma of prostate Introduction The grading system for prostatic adenocarcino- ma, developed by Gleason, has a strong prognos- tic value. The primary and secondary patterns are combined to give a Gleason score or sum. When only a minute focus of tumor is present in the specimen, the score is determined by dou- bling the number of Gleason pattern.(1) It has been claimed that Gleason score in biopsy spec- imen correlates with prostatectomy Gleason score and in combination with pretreatment Received June 2004 Accepted November 2004 *Corresponding author: Department of Pathology, Shariati hospital, Tehran University of Medical Sciences, Tehran, Iran. Tel: ++98 21 8490-2159. E-mail: tavangar@ams.ac.ir Tavangar et al 247 serum prostate-specific antigen (PSA) and digital rectal examination results, it can predict tumor stage and lymph node metastasis.(2) There are studies in the literature that have specifically cor- related needle biopsy and prostatectomy Gleason scores.(3-6) In many of these studies it has been noted that when in biopsy specimen, one encoun- ters a low-grade tumor, in a notable percent of cases the Gleason score will be higher in prosta- tectomy specimen. Thus, Gleason grading of a seemingly low-grade tumor in biopsy specimens may have unwanted effects on management of such patients. The aim of this study was to inves- tigate the correlation between Gleason score of biopsy and prostatectomy specimens. Materials and Methods Between 2000 and 2003, consecutive paired biopsy and prostatectomy specimens from 111 cases of prostatic adenocarcinoma, which were diagnosed by prostatic needle biopsy and had undergone radical prostatectomy in follow-up, were selected. Patients who had undergone neoadjuvant therapy as radiotherapy or androgen- deprivation therapy were excluded from the study. All biopsy specimens had been taken by 18- gauge needle, mostly under the guide of ultra- sonography, but the number of cores was varying between 4 and 10, because of different clinical experience of the urologists. The primary and secondary Gleason patterns and final Gleason scores of paired biopsy and prostatectomy (mini- mum of three slides per patient) were determined separately, blindly and without matching of paired samples. The analysis of agreement between biopsy and prostatectomy Gleason scores was based on individual scores and after assign- ment to one of the four groups defined as Gleason scores of 2 to 4, 5 to 6, 7 and 8 to 10. Correlation between Gleason scores of biopsy and prostatectomy specimens was analyzed by calcu- lating the coefficient of agreement (Kappa) and Pearson's correlation coefficient using SPSS 11.5 software. Accuracy of biopsy was also evaluated by determination of sensitivity, specificity, and positive and negative predictive values. Results Median age of the patients was 62 ± 10.6 (range 39 to 89) years. The most prevalent score was 6 (20.7%) in biopsy specimens and 7 (23.4%) in prostatectomy specimens (tables 1, 2). There was no score 10 tumor in any of biopsy or surgi- cal specimens. Most of the tumors in biopsy spec- imens were in the first grading group (low-grade, Gleason score 2 to 4) and most of the tumors in prostatectomy specimens were in the second group (medium-grade, Gleason score 5 to 6). The correlation between the Gleason scores of biopsy and prostatectomy is shown in Table 1. The Gleason scores were similar in 47.7%, and dif- fered by 1 point in 18% of cases. Overall, 45% were undergraded and 7.2% overgraded. Considering a maximal difference of one number as a desirable correlation, in 65.7% of the cases correlation was seen between biopsy and prosta- tectomy specimens. The most undergrading cases (84.2%) was observed in first group (Gleason score 2 to 4) and the most overgrading cases was seen in the last group (Gleason score 8 to 10). Kappa analysis yielded a value of 0.392 and Pearson's R was measured as 0.535 (table 1), cor- responding to a moderate agreement beyond chance and relative direct correlation between the biopsy and prostatectomy specimens. After grouping, the same analysis was done for the Gleason score group assignments (table 2). In this instance 55.8% of cases remained within the same group, 37.8% were undergraded and 6.3% were overgraded. Kappa and Pearson's R yielded values of 0.419 and 0.514 respectively. The accu- racy based on these group assignments is given in table 3. The sensitivity and positive predictive value for a biopsy Gleason score of 2 to 4 (low- grade carcinoma in biopsy specimen) was 90.9% and 26.3%, respectively, while for Gleason score TABLE 1. Correlation of biopsy and prostatectomy Gleason scores Pearson's R = 0.535, Kappa = 0.392 (P <0.0001) � ����� � �� �� �� �� �� �� �� ��� ����� � �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� ��� �� �� �� �� �� �� �� �� �� �� ��� �� �� �� �� �� �� �� �� �� �� ��� �� �� �� �� �� � � �� �� �� �� ��� �� �� �� �� �� �� ��� �� �� �� ��� �� �� �� �� �� �� �� �� �� �� ��� �� �� �� �� �� �� �� �� �� �� � � �� � �� � ��� �� �� �� �� �� �� �� �� �� �� � ������ �� �� �� ��� � � ��� ��� ��� �� ���� TABLE 2. Correlation of biopsy and prostatectomy Gleason scores by group assignment Pearson's R = 0.514, Kappa = 0.419 (P <0.0001) � � ����������� � � � ������ �� � �� �� ������ �� ��� ��� � � � � � �� ��� �� � ��� �� �� �� ��� �� ��� � �� � �� � �� �� �� �� ��� ��� � ������ ��� ��� ��� � � ���� Prostate Needle Biopsy and Radical Prostatectomy Gleason Gradings248 of 8 to 10 (high-grade carcinoma in biopsy) was 48.6% and 85%, respectively. There is clear evi- dence that more well-differentiated cancers have a higher frequency of being underscored and the poorly differentiated cancers being overscored in biopsy specimens. The correlation between the biopsy Gleason score and surgical Gleason score is shown in figure 1. The relationship between these, in the sense that well-differentiated can- cers are consistently undergraded and poorly dif- ferentiated cancers are consistently overgraded, is well fit by a linear regression (r2 = 0.29, P = 0.0001). Discussion Gleason grading system is important in deter- mination of prognosis and management of prosta- tic adenocarcinoma.(7) Gleason score in associa- tion with pretreatment serum PSA level and result of digital rectal examination predicts tumor stage and existence of lymph node metas- tasis.(8) Consequently, it is necessary to deter- mine the accuracy of needle biopsy scoring and correlation of this score with the one assigned to radical prostatectomy specimens. There are some studies in the literature comparing Gleason scores of biopsy and prostatectomy, in most of which it has been indicated that in some cases, especially when one encounters a low-grade tumor in biopsy, the assigned score underesti- mates the final score in the prostatectomy speci- men and contrarily needle biopsy scoring overes- timates prostatectomy scores to some extent in high-grade tumors.(3-6,8-13) Pearson's correlation coefficient and Kappa coefficient of agreement were calculated as 0.535 and 0.392, respectively, implying a moderate direct relationship between biopsy and prostatectomy Gleason scores (tables 1,2). The relationship between the biopsy Gleason score, in the sense that well-differentiated can- cers are consistently undergraded and poorly-dif- ferentiated cancers overgraded is well shown in figure 1. In low-grade tumors (Gleason grade 2 to 4) 84.2% of cases were undergraded. In com- parison, only 5% undergrading was found in high- grade tumors. On the other hand, needle biopsy Gleason scores of 20% of high-grade tumors were overestimated, while no overgrading was observed in low-grade tumors. As an index of accuracy, the positive predictive value of Gleason scoring in biopsy was only 26.3% in low-grade tumors and reached to 85% in high-grade tumors, implying insufficient accuracy in low-grade tumors (table 3). Different factors have been suggested as the reasons of this significant undergrading of low- grade tumors in biopsy specimens. Its consisten- cy in different studies implies that it is more a systematic bias toward undergrading, rather than an error in pathologic interpretation. Gleason has proposed that the undergrading may be due to several sources including reluctance of patholo- gists to characterize a small amount of high- grade tumor in an otherwise low-grade back- ground.(4) Other factors may contribute to the discrepancies between Gleason score of biopsies and surgical specimens as the amount of cancer- ous tissue present within biopsy material and sampling effects.(4) To determine whether the amount of cancerous tissue in the biopsy speci- men is responsible for the Gleason score differ- ence between the prostatectomy and biopsy, a correlation analysis has been performed by King,(4) Bostwick,(9) and Steinberg;(3) none of them have found any significant correlation. Since prostate cancer is often multifocal, with a TABLE 3. Accuracy of biopsy Gleason score in predicting final surgical Gleason group ���������� � ������������������������� ���������� � !� "��#����� $� !� "��#������ $� ��� ��� ������ ���� ���� � ����� �� ��� ������ ���� ������ ����� �� ������ ���� ������ ����� �� ���� ���� � � �� ������ ��� FIG. 1. Relationship between the biopsy Gleason score and surgical Gleason score. Error bars represent the 95% confi- dence interval about the mean. The number of cases in different groups is indicated above error bars. Regression is indicated by the solid line (r2 = 0.29, P = 0.0001). The dashed line represents perfect correlation. � �� �� � �� �� � �� ����������� � �� � �� �� �� �� �� �� �� �� � ����� � ������ � ������ � ��� � � ������ � ������ � ������ � ������ � � � �� � �� � ��� � � �� �� � � � � Tavangar et al 249 heterogeneous population of tumor cells, a cer- tain degree of sampling error is inevitable. This may result in sampling an area that consists of more high-grade or low-grade tumor samples than the actual tumor. It has been suggested that to overcome sam- pling error one must either perform a directed biopsy (if there is an ultrasound-visible lesion) or increase the number of biopsies obtained. Some studies suggest that sampling error might be sig- nificantly reduced by obtaining more biop- sies.(12,14) Some authors propose a modification to the Gleason system to include "tertiary" or the third most prevalent pattern in the scoring,(15) but King(4) argues that this modification may even increase the error of sampling. Also a rou- tine consensus approach to pathologic evaluation of prostate adenocarcinoma seems useful. Conclusion According to our findings, there is a moderate direct linear relationship between scores in biop- sy and prostatectomy specimens. But there is a high probability of underestimation of real Gleason score of the radical prostatectomy speci- men in low-grade tumors. Pathologists and urolo- gists must consider the phenomenon of under- grading in reporting prostate specimens and managing patients. It must be emphasized that radical therapies for localized prostate adenocarcinoma are sometimes determined or excluded from consideration on the basis of the biopsy Gleason score. Now the differences between the histological grade in biop- sies and surgical specimens are being under- stood. Therefore, staging of organ confined prostate cancer, when based on biopsy grading, should include the likelihood of histological over- estimation in the surgical specimen. References 1. Deshmukh N, Foster CS. Grading prostate cancer. In: Foster CS, Bostwick DG, editors. Pathology of prostate.1st ed. Philadelphia: WB Saunders; 1998. p.191-227. 2. Epstein JI, Partin AW, Sauvageot J, Walsh PC. Prediction of progression following radical prostatecto- my: a multivariate analysis of 721 men with long-term follow-up. Am J Surg Pathol. 1996;20:.286-92. 3. Steinberg DM, Sauvageot J, Piantadosi S, Epstein JI. Correlation of prostate needle biopsy and radical prosta- tectomy Gleason grade in academic and community set- tings. Am J Surg Pathol 1997;21:566-76. 4. King CR. Patterns of prostate cancer biopsy grading, trends and implications. Int J Cancer. 2000;90:305-11. 5. Ruijter E, van Leenders G, Miller G, Debruyne F, van de Kaa C. Erros in histological grading by prostatic needle biopsy specimens: frequency and predisposing factors. J Pathol. 2000;192:229-33. 6. Prost J, Gros N, Bastide C, Bladou F, Serment G, Rossi D. Correlation between Gleason score of prostatic biop- sies and one of the radical prostatectomy specimens. Prog Urol. 2000;11:45-8. 7. Gleason D. Classification of prostate carcinoma. Cancer Chemother Rep. 1996;50:125-8. 8. Rubin MA, Dunn R, Kambham N, Misick CP, O'Toole KM. Should a Gleason grade be assigned to a minute focus of carcinoma on needle biopsy? Am J Surg Pathol. 2000;24:1634-40. 9. Bostwick DG. Gleason grading of prostate needle biop- sies. Correlation with grade in 316 matched prostatec- tomies. Am J Surg Pathol. 1994;18:796-803. 10. Epstein JI. Gleason score 2-4 adenocarcinoma of prostate on needle biopsy. Am J Surg Pathol. 2000;24:477-8. 11. De la Taille A, Viellefond A, Berger N, et al. Evaluation of interobserver reproducibility of Gleason grading of prostatic adenocarcinoma using tissue microarrays. Hum Pathol. 2003;34:444-9. 12. San Francisco IF, De Wolf WC, Rosen S, Upton M, Olumi AF. Extended prostate needle biopsy improves concordance of Gleason grading between prostate needle biopsy and radical prostatectomy. Urology. 2003;169:136-40. 13. Djavan B, Kadesky K, Klopukh B, Marberger M, Roehrborn CG. Gleason scores from prostatic biopsies obtained with 18-gauge biopsy needles poorly predict Gleason scores of radical prostatectomy specimens. Eur Urol. 1998;33:261-70. 14. Thickman D, Speers WC, Philpott PJ, Shapiro H. Effect of the number of core biopsies of the prostate on predict- ing Gleason score of prostate cancer. J Urol. 1996;156:110-3. 15. Pan CC, Potter SR, Partin AW, Epstein JI. The prognos- tic significance of tertiary Gleason patterns of higher grade in radical prostatectomy specimens: a proposal to modify the Gleason grading system. Am J Surg Pathol. 2000;24:563-9.